In a conference held in Chicago during October 1995, a working group of the National Institute of Aging (NIA) and the Alzheimer's Association (AlzA) drafted consensus recommendations on research and clinical applications of APOE genetic susceptibility testing for Alzheimer's disease (AD). The NIA/AlzA Working Group concluded that in considering future applications of APOE genotyping and other knowledge that has been gained about the genetic basis of AD, the interests of AD patients and their family members must be held paramount. The group acknowledged that a robust association exists between possession of the APOE epsilon 4 allele and the risk of late-onset AD and cited evidence that this allele is more strongly associated with AD than any other form of dementia. They recommended against the use of APOE genotyping to predict the-future development of AD in asymptomatic individuals at this time, and warned against the use of the test in isolation as the sole means for diagnosing AD. The group endorsed the concept of discretionary use of APOE genotyping as an adjunct to other AD diagnostic procedures. However, routine clinical use of the test for this purpose was not recommended at this time. Physicians were advised to weigh any potential benefits of testing against the possibility that genotype disclosure could adversely affect the insurability, employability, and social standing of AD patients and their family members. Adequate provisions for pre-test and post-test counseling and psychosocial support were advised for all future clinical and research applications of APOE genotyping. The group called for the development of improved protocols for AD genetic counseling as well as supplemental measures to assure genetic privacy for AD patients and their family members.